Effectiveness and Renal Safety of Tenofovir Alafenamide Fumarate Among Chronic Hepatitis B Patients

Real-World Study

Mina S. Farag; Scott Fung; Edward Tam; Karen Doucette; Alexander Wong; Alnoor Ramji; Brian Conway; Curtis Cooper; Keith Tsoi; Philip Wong; Giada Sebastiani; Mayur Brahmania; Sarah Haylock-Jacobs; Carla S. Coffin; Bettina E. Hansen; Harry L.A. Janssen


J Viral Hepat. 2021;28(6):942-950. 

In This Article

Materials and Methods

Population and Study Design

Chronic hepatitis B patients receiving TAF were included from eleven institutions across Canada as part of the Canadian Hepatitis B Network (CanHepB). Data were collected from participating sites through an online web-based portal following local chart review at each site from January 2015 to December 2019. Patients were excluded if enrolled in a clinical trial did not complete at least three visits among NA-experienced patients (pre-TAF visit, baseline, TAF visit) or two visits for NA-naïve patients (baseline, TAF visit). In addition, patients co-infected with hepatitis C virus (HCV), hepatitis delta virus (HDV) or human immunodeficiency virus (HIV) were excluded. This study was conducted in accordance with the 1975 Declaration of Helsinki. All data received were anonymous and collected under the protocol approved by the institutional research boards of participating centres, as per local regulations and medical legal agreements for data sharing between sites within the Canadian Hepatitis B Network (Can Hep B).

Data Collection

Baseline was defined as the starting date of TAF treatment. Demographic and clinical information were available at baseline. Renal comorbidity was defined by baseline presence of either diabetes mellitus, hypertension and metabolic syndrome (according to The National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III)).[15] Also, the following laboratory values were serially assessed at baseline and follow-up visits: HBV DNA, hepatitis B e antigen (HBeAg), quantitative Hepatitis B surface antigen (HBsAg), alanine aminotransferase (ALT), aspartate transferase (AST), liver stiffness measurement using transient elastography (TE, FibroScan®) and lipid profile. Renal function was measured by serum creatinine, estimated glomerular filtration rate (eGFR) as per Cockcroft-Gault and serum phosphate. HBV DNA testing was performed in provincial diagnostic laboratories using commercial real-time polymerase chain reaction (PCR) (i.e. TaqMan®) assays (i.e. Abbott Architect, sensitivity <10 IU/mL and Roche Cobas Amplicor, sensitivity <20 IU/mL; 1 IU/mL equivalent to ~5 virus genome copies/mL). HBV serological testing (HBsAg, HBeAg) was performed using standard commercial immunoassays (i.e. Abbott Architect) in provincial laboratories and accredited hospital diagnostic laboratories. Routine biochemical and haematological tests were performed locally at participating centres. Serum ALT levels were standardized against the upper limits of normal recommended by AASLD (the American Association for the Study of Liver Diseases) and CASL (Canadian Association for the Study of the Liver), ≤35 U/L for males and ≤25 U/L for females.[4,16] Liver cirrhosis was defined as liver stiffness >11 Kpa or as diagnosed by liver biopsy.

Patients who started TAF were either categorized as NA naïve or experienced. Patients categorized as NA-experienced were further subcategorized according to the specific NA used before switching to TAF. Data were obtained for visits up to 160 weeks before switching to TAF and 160 weeks thereafter. Additionally, NA-naïve patients were followed-up for up to 160 weeks. Follow-up visits occurred during standard of care visits based on physician discretion every three to six months. In both groups (NA-naïve and experienced), patients were classified according to their chronic kidney disease (CKD) stages: eGFR [stage 1 (≥90), 2 (60–89), 3a (45–59), ≥3b (< 45)] at baseline. Analysis of eGFR changes is based on current renal guidelines.[17,18]

Statistical Analyses

Continuous variables were presented as mean (standard deviation) or median (interquartile range). Non-normally distributed variables were log-transformed or categorized. Patient characteristics were compared using Student's t-tests, Mann–Whitney tests or Wilcoxon signed-rank tests for continuous variables and chi-square tests or Fisher's exact tests for categorical variables. Exploratory subgroup analyses based on baseline characteristics were performed. Kaplan–Meier curves were generated to analyse the incidence of achieving undetectable HBV DNA after starting TAF.

The McNemar test was used to compare decline (yes/no) in eGFR before and after TAF. To correct for within-subject variability, mixed linear regression models with random intercept and slopes were applied to estimate the trend of eGFR before and after TAF. The trend model of eGFR before TAF administration, with week=0 is start of TAF, was as follows: eGFR=intercept + slope1 * week, and after TAF administration eGFR=intercept + slope2 * week, with slope2 = slope1+change in slope1, week >0). The overall change in slope before vs after TAF was estimated. The analysis was repeated for specific subgroups at the switch to TAF, such as CKD stages, presence of renal comorbidity and pre-TAF NA experience. The above trend analysis was also applied to the changes in HBsAg, serum phosphate and ALT before and after TAF.

A P-value less than 0.05 was considered statistically significant. All analyses were two-sided and were performed using IBM SPSS Statistics, version 25.0 (IBM Corp., Armonk, NY, USA) and SAS version 11.2 (SAS Institute Inc., Cary, NC, USA).